Database queries, in particular, event-driven continuous queries, are useful for many pervasive computing applications, such as video surveillance. In order to enable these applications, we have developed a pervasive ...
详细信息
ISBN:
(纸本)0769523315
Database queries, in particular, event-driven continuous queries, are useful for many pervasive computing applications, such as video surveillance. In order to enable these applications, we have developed a pervasive query processing framework called Aorta. Unlike traditional database systems, a pervasive query processor requires systems support for managing a large number of networked, heterogeneous devices. In this paper, we present the communication, synchronization, and scheduling mechanisms in Aorta. Even though these techniques have their roots in distributed and parallel systems, we show how these techniques are customized and applied for pervasive query processing. In essence, communication between heterogeneous devices enables network data independence, synchronization on devices protects action atomicity, and scheduling works for adaptive, cost-based multi-query optimization. We have conducted empirical studies on our prototype as well as simulation studies to evaluate the system performance.
This paper presents a coherence protocol for recoverable distributed Shared Memory, (DSM) systems with causally consistent read-write objects. It uses independent checkpointing tightly integrated with coherence operat...
详细信息
ISBN:
(纸本)1892512459
This paper presents a coherence protocol for recoverable distributed Shared Memory, (DSM) systems with causally consistent read-write objects. It uses independent checkpointing tightly integrated with coherence operations. That integration results in high availability of shared objects and ensures fast restoration of the consistent state of DSM in spite of multiple node failures, introducing little overhead. Moreover, in case of network partitioning, the extended protocol ensures that all the processes in a majority partition of the DSM system can continuously access all the objects. The integration can, in some circumstances, also improve the efficiency of the coherence operations.
Increasingly, services need to interact with a wide spectrum of devices with varying networking capabilities. Services hosted on a messaging infrastructure need to optimally utilize and exploit the conditions that exi...
详细信息
ISBN:
(纸本)1892512459
Increasingly, services need to interact with a wide spectrum of devices with varying networking capabilities. Services hosted on a messaging infrastructure need to optimally utilize and exploit the conditions that exist within the local networks. The messaging infrastructure must manage the communication between external resources, services and clients to achieve the highest possible system performance and reliability. In this paper we suggest that a transport framework needs to be incorporated into the messaging infrastructure hosting the services. We base our investigations in the context Of the NaradaBrokering system.
In this paper, we design and analyze an election algorithm, based on the Bully algorithm, in synchronous distributed systems. We show that the Bully algorithm, when using Failure Detector, is more effectively implemen...
详细信息
ISBN:
(纸本)1601320841
In this paper, we design and analyze an election algorithm, based on the Bully algorithm, in synchronous distributed systems. We show that the Bully algorithm, when using Failure Detector, is more effectively implemented than the classic Bully algorithm for synchronous distributed systems.
An integrated management system for large-scale simulations saves the output data simulation results without considering the amount of empty space on the disks in the data servers. The present problem is inefficient o...
详细信息
ISBN:
(纸本)1892512459
An integrated management system for large-scale simulations saves the output data simulation results without considering the amount of empty space on the disks in the data servers. The present problem is inefficient operations on the data servers because each the disk sizes on the data server are unbalanced. We design and implement a prototype parallel I/O control agent for large-scale simulation data servers which supervises the data servers, makes suggestions to users and moves data flies between disks to keep disk sizes well-balanced.
Reproducibility of the execution of scientific applications on parallel and distributed systems is a growing concern, underlying the trustworthiness of the experiments and the conclusions derived from experiments. Dyn...
详细信息
ISBN:
(纸本)9780769561493
Reproducibility of the execution of scientific applications on parallel and distributed systems is a growing concern, underlying the trustworthiness of the experiments and the conclusions derived from experiments. Dynamic loop scheduling (DLS) techniques are an effective approach towards performance improvement of scientific applications via load balancing. These techniques address algorithmic and systemic sources of load imbalance by dynamically assigning tasks to processing elements. The DLS techniques have demonstrated their effectiveness when applied in real applications. Complementing native experiments, simulation is a powerful tool for studying the behavior of parallel and distributedapplications. In earlier work, the scalability [1], robustness [2], and resilience [3] of the DLS techniques were investigated using the MSG interface of the SimGrid simulation framework [4]. The present work complements the earlier work and concentrates on the verification via reproducibility of the implementation of the DLS techniques in SimGrid-MSG. This work describes the challenges of verifying the performance of using DLS techniques in earlier implementations of scientific applications. The verification is performed via reproducibility of simulations based on SimGrid-MSG. To simulate experiments selected from earlier literature, the reproduction process begins by extracting the information needed from the earlier literature and converting it into the input required by SimGrid- MSG. The reproducibility study is carried out by comparing the performance of SimGrid-MSG-based experiments with those reported in two selected publications in which the DLS techniques were originally proposed. While the reproduction was not successful for experiments from one of the selected publications, it was successful for experiments from the other. This successful reproduction implies the verification of the DLS implementation in SimGrid- MSG for the considered applications and systems, and th
In this paper we introduce methods for reconfiguring an omega network of identical processing elements (PEs) so that a reliable operation is possible in the presence of multiple faulty PEs. The proposed approach ensur...
详细信息
ISBN:
(纸本)1892512459
In this paper we introduce methods for reconfiguring an omega network of identical processing elements (PEs) so that a reliable operation is possible in the presence of multiple faulty PEs. The proposed approach ensures the maximum utilization of the spare processing elements and is superior compared to the known methods, because it can tolerate multiple PE failures on an omega-route. The network is capable of dynamically reconfiguring itself using (a) simple equations along with a set of well-defined switching rules or (b) a max-flow based algorithm. We describe the required hardware and the reconfiguration methods and we present the application of these methods in typical failure cases.
This paper explores object-oriented distributed transaction scheduling in workflow management systems. The purpose of this paper is to present efficient protocol of the transaction scheduling at the workflow managemen...
详细信息
ISBN:
(纸本)1892512416
This paper explores object-oriented distributed transaction scheduling in workflow management systems. The purpose of this paper is to present efficient protocol of the transaction scheduling at the workflow management system, using message queues of five types and thread architecture. Especially, focused is the protocol about scheduling of subtransactions, which is divided from the arrived transactions, at schedule queue. The message queuing system is designed for synchronization between subtransactions. The thread architecture is exploited for parallel transaction scheduling. We will show a simulation to demonstrate efficiency of our scheduling protocol. We will also present parallel transaction processing with the aid of load balancing in the workflow management system.
GPUs have been used to accelerate different data parallelapplications. The challenge consists in using GPUs to accelerate stream processingapplications. Our goal is to investigate and evaluate whether stream paralle...
详细信息
ISBN:
(纸本)9781728116440
GPUs have been used to accelerate different data parallelapplications. The challenge consists in using GPUs to accelerate stream processingapplications. Our goal is to investigate and evaluate whether stream parallelapplications may benefit from parallel execution on both CPU and GPU cores. In this paper, we introduce new parallel algorithms for the Lempel-Ziv-Storer-Szymanski (LZSS) data compression application. We implemented the algorithms targeting both CPUs and GPUs. GPUs have been used with CUDA and OpenCL to exploit inner algorithm data parallelism. Outer stream parallelism has been exploited using CPU cores through SPar. The parallel implementation of LZSS achieved 135 fold speedup using a multi-core CPU and two GPUs. We also observed speedups in applications where we were not expecting to get it using the same combine data-stream parallel exploitation techniques.
ParJava integrated environment supporting the development and maintenance of data parallel Java-programs is discussed When a parallel program is developed it is necessary to assure not only its correctness, but also i...
详细信息
ISBN:
(纸本)1892512459
ParJava integrated environment supporting the development and maintenance of data parallel Java-programs is discussed When a parallel program is developed it is necessary to assure not only its correctness, but also its efficiency and scalability. For this purpose it is useful to know some dynamic properties of the program (profiles, traces, slices, etc). This information may help to modify the program in order to improve its features. ParJava provides a collection of tools, which help to determine and to improve parallel program properties. These properties are used during fast execution of SPMD-program allowing to estimate its expected execution time and scalability. It seems to be very convenient facility that allows to estimate limits of scalability and thus to determine needed number of processor nodes.
暂无评论